Visual Planning on CrossTask
38.45Success Rate (SR)ViterbiPlanNet
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ViterbiPlanNetParams (M)=5.572026.03 | 38.45 | 63.07 | 83.89 | |
| SCHEMAParams (M)=6.132026.03 | 37.24 | 62.69 | 83.94 | |
| PlanLLMParams (M)=384.942026.03 | 36.84 | 61.56 | 83.23 | |
| PDPPParams (M)=41.872026.03 | 36.73 | 61.96 | 83.2 | |
| KEPPParams (M)=42.182026.03 | 34.93 | 60.34 | 82.67 | |
| Gemini 2.5 ProParams (M)=>100,0002026.03 | 29.18 | 57.9 | 81.48 | |
| Qwen2.5-32BParams (M)=32,0002026.03 | 25.14 | 56.1 | 80.92 | |
| ViterbiPlanNetParams (M)=5.602026.03 | 24.64 | 57 | 81.18 | |
| SCHEMAParams (M)=6.132026.03 | 24.18 | 57.02 | 81.46 | |
| Qwen3-30BParams (M)=30,0002026.03 | 23.37 | 55.96 | 81.16 | |
| Qwen3-30B + PKGParams (M)=30,0002026.03 | 23.31 | 56.15 | 81.06 | |
| PlanLLMParams (M)=384.942026.03 | 22.91 | 55.29 | 81.03 | |
| PKG beam searchParams (M)=41.872026.03 | 22.38 | 55.74 | 80.92 | |
| KEPPParams (M)=42.182026.03 | 22.34 | 55.24 | 80.58 | |
| PDPPParams (M)=41.872026.03 | 21.47 | 55.66 | 80.68 | |
| Gemini 2.5 ProParams (M)=>100,0002026.03 | 14 | 51.33 | 78.58 | |
| Qwen2.5-VL-32BParams (M)=32,0002026.03 | 11.48 | 36.35 | 69.52 | |
| Qwen3-30B + PKGParams (M)=30,0002026.03 | 10.96 | 48.77 | 77.48 | |
| Qwen3-30BParams (M)=30,0002026.03 | 10.59 | 49.06 | 78.03 | |
| PKG beam searchParams (M)=41.872026.03 | 9.3 | 47.65 | 78.25 | |
| Qwen2.5-32BParams (M)=32,0002026.03 | 9.22 | 46.32 | 76.15 | |
| Qwen2.5-VL-32BParams (M)=32,0002026.03 | 5.56 | 31.22 | 66.31 |